I Biochemical Basis of Thyroid Hormone Action in the Heart WOLFGANGH. DILLMANN,
M.D., San Diego, Cafifornia
Thyroid hormone-induced changes in cardiac function have been recognized for over 150 years; however, the biochemical basis of triiodothyronine (T3) action in the heart hA~ been intensely investigated only during the last two decades. T3=induced changes in cardiac function can result from direct or indirect T3 effects. Direct T3 effects result from T3 action in the heart itself and are mediated by nuclear or extranuclear mechanisms. Extranuclear T3 effects, which occur independent of nuclear T3 receptor binding and increases in protein synthesis, influence primarily the transport of amino acids, sugars, and calcium across the cell membrane. Nuclear T3 effects are mediated by the binding of T3 to specific nuclear receptor proteins, which results in increased transcription of T3-responsive cardiac genes. The T3 receptor is a member of the ligandactivated trRnseription factor family and is encoded by cellular erythroblastosis A (c-erb A) genes. The c-erb A protein is the cellular homologue of the viral erythroblastosis A (v-erb A) protein, which causes red cell leukemi~ in chickens. Currently, three T3-binding isoforms of the c-erb protein and two non-T3-binding nuclear proteins that exert positive and negative effects on T3-responsive cardiac genes have been identified. T3 increases the heart trRn~cription of the myosin heavy cheln (MHC) a gene and decreases the transcription of the MHC ~ gene, leading to an increase of myosin VI and a decrease in myosin V3 isoenzymes. Myosin V], which is composed of two MHC ~, has a higher myosin ATPase activity than myosin V3, which cont~i-~ two MHC ~. The globular head of myosin V1, with its higher ATPase activity, leads to a more rapid movement of the glob, i~r head of myosin along the thin filament, resulting in an increased velocity of contraction. T3 also leads to an increase in the speed of diastolic relaxation, which is caused by the more efficient pumping of the calcium ATPase of the sarcoplasmic reticulum (SR). This T3 effect results from T3-induced increases in the level of the mRNA coding for the SR calcium ATPase protein, leading to an increased number of calcium ATPase pump units in the SR. Overall, thyroid hormone leads to an increase in ATP consumption in the heart. In addition, less chemical energy of ATP is used for contractile purposes and more of it goes toward heat production, which causes a decreased efficiency of the contractile process in the hyperthyroid heart. From the Department of Medicine, Division of Endocrinology and Metabolism, University of California, San Diego, San Diego, California. This work was supported by Grant HL25022 from the National Institutes of Health, Bethesda, Maryland, and was presented in part at the 64th Annual Meeting of the American Thyroid Association, San Francisco, California. Requests for reprints should be addressed to W.H. Dillmann, M.D., University of California at San Diego Medical Center, 225 Dickinson Street (H-811-C), San Diego, California 92103.
626
June 1990 The American Journalof Medicine
Volume 88
he heart is a major target organ for thyroid horT mone action, and marked changes occur in cardiac function in patients with hypothyroidism or hyperthyroidism [1]. This was first recognized in 1785 by Caleb Parry [2], a physician from the English resort town of Bath, who noted an association between swelling of the thyroid area and heart failure. He stated "there is one malady which I have in five cases seen coincidence which appears to be an enlargement of the heart. The malady to which I elude is enlargement of the thyroid gland." Fifty years later Robert Graves [3] described three cases of violent and long continued palpitation in women with thyrotoxicosis. Many important studies related to physiologic alterations induced by changes in the thyroid status were conducted subsequently; however, a detailed elucidation of the biochemical basis of thyroid hormone effects in the heart occurred only during the last 2 decades. It was also appreciated that thyroid status-induced changes in cardiac function can result from direct cardiac effects of thyroid hormone, that is, action of triiodothyronine (T3) itself in the cell of the heart versus indirect T3 effects, which are mediated by alterations occurring in the periphery resulting in hemodynamic changes that lead to secondary alterations in cardiac function. This discussion will be focused on direct effects of thyroid hormone on the heart. In considering thyroid hormone effects on cardiac tissue, it is important to emphasize that cardiac myocytes, which due to their large size contain the majority of cardiac protein and ribonucleic acids, constitute only one-third of the total cells that make up the myocardium [4]. The majority of cardiac cells are constituted by fibroblasts, smooth muscle cells, endothelial cells, and other cell types. Thyroid hormone-responsive proteins that have been investigated are primarily of myocytic origin, and thyroid hormone influences on nonmyocytic cardiac cells have not been investigated in detail. The first step in the sequence of events leading to thyroid hormone action in cardiac cells involves the transport of thyroid hormone across the plasma membrane or sarcolemma of cardiac myocytes, and subsequently transport across the nuclear membranes into the cell nucleus occurs. Recent findings by several groups of investigators indicate that an energy-requiring stereospecific transport step of thyroid hormone and its an~ogues occurs at the level of the plasma and nuclear membrane [5-8]. The stereospecific transport may account for discrepancies between nuclear thyroid hormone receptor binding and biologic activity. It has, for example, been noted that the binding of the LT3 isomer for the nuclear T3 receptor protein is only 15% to 20% higher than that of the D-T3 isomer [9]. In contrast, the biologic activity of D-T3 is only one third of L-T3. An explanation for this difference in receptor binding affinity and biologic activity may be provided by the preferential transport of L-T3 versus D-T3 across the plasma and nuclear membrane in the heart
T3 ACTION IN THE HEART / DILLMANN
[7]. Similar observations have been made in a rat myoblast cell line [8]. Once thyroid hormones enter cardiac cells, they mediate their effects through either a nuclear mechanism by binding to specific nuclear T3 receptors and stimulation of protein synthesis or by socalled extranuclear effects, which occur independent of binding to specific nuclear T3 receptor proteins and do not require ongoing protein synthesis. Extranuclear effects resulting in rapid stimulation of amino acid and sugar transport have been demonstrated [10]. In addition, thyroid hormone seems to exert an extranuclear effect on the sarcolemmal calcium ATPase, which results in increased calcium efflux from myocytes [11]. The potency of specific thyroid hormone analogues mediating this effect does not correspond to that observed for nuclear action. For example, T3 is almost as potent as thyroxine (T4) in stimulating sarcolemmal calcium ATPase [11]. In addition, direct effects of thyroid hormone on mitochondrial oxidative phosphorylation have been postulated [12]. Thyroid hormone effects on protein synthesis result in enhanced formation of total cardiac proteins [13] and, in addition, increases in the synthesis rate of the specific proteins that exceed the general increase in protein synthesis induced by thyroid hormone. For other specific proteins like myosin heavy chain (MHC) fl, the synthesis rate may decrease. T3-induced changes in the formation of specific proteins can result from alterations at the following steps. Thyroid hormone can (1) influence the formation of specific proteins by increasing the transcription of a given gene; (2) control how the primary transcript is spliced or otherwise processed; (3) select which mRNA species leave the nucleus and are exported to the cytoplasm; (4) selectively degrade certain mRNAs in the cytoplasm, influencing the half-life of the mRNA; (5) determine which mRNA species in the cytoplasm translate p r e f e r e n t i a l l y by ribosomes, l e a d i n g to translational control; and (6) selectively activate or degrade specific protein molecules after they have been made, leading to protein activity control. Thyroid hormone-induced changes in the formation of specific proteins appear to be controlled primarily at the steps regulating the transcription of specific genes and by influences on the stability of the mRNA in the cytoplasm. Thyroid hormone-induced changes in transcription are initiated by the binding of T3 to nuclear T3 receptors. Evidence for the existence of such receptors was first presented in 1972 and 1973 [14,15]. T3 receptor proteins are acidic nuclear proteins with a molecular weight of roughly 50,000 dalton (d) and bind thyroid hormone with high affinity. Because T3 is bound to this receptor protein with a higher affinity than T4, this thyroid hormone analogue has a higher biologic activity [16]. The following model for nuclear T3 effects was therefore developed. T3 enters the cell and the nucleus and is bound to the specific T3 receptor protein that is always present in association with chromatin. By binding of T3 to these receptor proteins, the transcription of specific genes is activated and an ificreasing amount of specific mRNA is formed. Attempts to purify T3 receptor proteins by conventional methods were unsuccessful because the protein was present at a low level (5,000 to 10,000 receptors/cell). In addition, the protein is labeled and difficult to puri-
TABLE I c.erb A Nuclear Proteins (T3 Receptors)*
Gene
Isoforms
c-erb Aa /
Protein
I
V/
I
I ,////I
N
~ o ~ =
T3 Effect
E
C
I' ~, GH
E
E c-erb Aft__......_
N ,~2
[
I
t Pituitary
E c-erbAreverse
--
v-erbA
--
N I
C I"1////I E
N I
C I /.///I E
,~GH
N = amino-terminal portion of the protein; C = carboxy-terminal portion of the protein; E = E domain of the receptor protein where T3 binds;" = increase in gene transcription; = decrease in gene transcription; GH = growth hormone. * Different c-erb A nuclear proteins are listed in this table. They are encoded by two different genes, c-erb A~ and c-erbA/~. The splicing variants c-erb Ao~2 and c-erb A/~2 arise from the o~ gene and c-erb A~2 does not bind T3 in the E domain of the receptor protein. The c-erbAo~2protein inhibits T3 effects on growth hormone gene transcription.
fy away from other proteins in the 40,000 to 60,000-d molecular weight range, which are of much higher predominance. Further characterization and identification of the nuclear T3 receptors as members of the cellular erythroblastosis (c-erb) A proto-oncogene family did not occur until 15 years later [17,18]. The first step in this identification process was the recognition that the amino acid sequence of the DNA binding region of the glucocorticoid receptor had a 47% homology to the viral erythroblastosis (v-erb) A protein [19]. This protein is a product from an oncogene virus that in conjunction with the v-erb B protein induces a red cell leukemia in chickens. The cellular homologue of this protein c-erb A had a molecular weight of 50,000 d and had the characteristics of a hormone-binding nuclear protein. It did not bind cortisol or other steroid hormones but bound T3 with similar affinities to those of the naturally occurring T3 receptor. The nuclear T3 receptor was therefore identical to the c-erb A protein. In addition, it could be shown that several isoforms of the T3 receptor-c-erb A proteins occur (Table I). At least two different genes, one coding for a c-erb Aa T3 receptor isoform and a second gene coding for the cerb A~ isoform, have been initially identified [17,18]. The transcriptional product of the c-erb Aa gene is further processed into the c-erb Aalfr3 receptor, which binds T3 and has a positive effect on the transcription of the growth hormone gene, and an alternatively spliced variant c-erb Aa2 is formed in cells, which does not bind T3 [20-24]. This isoform inhibits the transcription of the growth hormone gene but does not seem to have an influence on the transcription of the MHC a gene. The c-erb AB transcript is processed into a f~l and a f12form. The c-erb A~2 variant occurs in the pituitary and cannot be found in the heart [25]. The DNA region that codes for the c-erb Aa gene is
June 1990 The American Journal of Medicine Volume 88
627
T3 ACTION IN THE HEART/ DILLMANN TABLE II T3-Responsive mRNA in the Heart Type
MHCo
MHC fl SR calcium ATPase Malic enzyme
Gene
mRNA
;
t
Protein
Activity
t My°s'nAzPaseI Myosin ATPase Calcium sequestration 1.
Sodium efflux 1. ATPase ANF Calcium channel fl Receptor
;
1.
1'
?
?
?
--
--
1.
Sodium excretion 1. Calcium influx ~' Sympathetic stimulus 1.
ANF = atrial natriuretic factor; 1' = increase in level; J, = decrease in level; NAD PH = nicotinamide adenine dinucleotide phosphate hydrogen..
also transcribed in the reverse direction, leading to the c-erb Aa reverse protein, which does not bind T3 [26,27]. This DNA region can therefore lead to three isoforms of the c-erb Aa T3 receptor protein: the ~1 form, which binds T3, the ~2 form, which does not bind T3, and the reverse isoform, which also does not bind to T3. In cardiac tissue, the mRNA coding for c-erb Aal, c-erb A~2, c-erb Aflb and c-erb A reverse can be documented [26,28]. The v-erb A proto-oncogene also has an inhibitory effect on the action of the c-erb A~I T3 receptor [29,30]. The different isoforms of the T3 receptor and the nuclear receptor for steroid hormones are from a family of ligand-activated transcriptional factors that show a significant degree of homology [31]. All hormone ligands contain a ring structure made up of six carbon atoms and are lipophilic and hydrophobic. The ligand binds the nuclear receptor protein at the carboxyterminal end. The receptor protein contains a DNA binding region (C region) in which these different receptors show a high degree of homology. The T3 receptor proteins bind with the C-region to areas in the DNA containing nucleotide consensus sequences that are localized upstream from the origin of transcription of T3-responsive genes. It is of interest to note that, for example, the growth hormone gene is thyroid hormone-responsive and also responds to glucocorticoids and to retinoic acid [32-36]. One could Speculate that during evolution, a primordial six-carbon ligand and a corresponding binding protein arose and were then modified into the specific receptors for thyroid hor: mone, glucocorticoids, retinoic acid, and other ligands. In addition, specific genes remained responsive to those ligands and their receptor proteins. The following model of T3 receptor action arises from recent findings. It appears that the T3 receptor is always associated with DNA in areas upstream from the transcription initiation site of thyroid hormone-responsive genes. The unoccupied T3 receptor acts as a transcriptional inhibitor, repressing the transcription of the thyroid hormone-responsive gene. Association of T3 with the carboxy-terminal E domain of the T3 receptor leads to a change in the confirmation of the receptor protein, which relieves the repressive function and allows the gene to be transcribed. Illustrative of how T3-induced changes in the tran-
628
scription of specific genes can result in contractile alterations are the T3 effect on the MHC ~ gene and the sarcoplasmic reticulum (SR) calcium ATPase gene. We will first consider alterations in MHC gene expression. It has been known for some time that increases in thyroid hormone levels lead to increases in the velocity of systolic contraction of the heart [37]. These alterations can be explained by alterations in myosin isoenzyme predominance. The myosin holoenzyme has a molecular weight of approximately 500,000 d and consists of two MHCs and four light chains [38]. Myosin V1, which predominates in the normal rat heart, consists of two MHC ~, whereas myosin V3 contains two MHC/3, and myosin V2 is a heterodimer of MHC ~ and MHC ~. Myosin V1 has markedly higher myosin ATPase activity than myosin V3. In the hypothyroid rat heart, myosin V3 predominates and myosin with low ATPase activity participates in the contractile process, leading to a decreased velocity of contraction in hypothyroid papillary muscle [39,40]. In contrast, thyroid hormone administration to hyperthyroid rats leads to increased expression of the MHC ~ gene, resulting in a markedly higher fraction of myosin isoenzymes being composed of two MHC a. The higher myosin ATPase activity of myosin V1 leads to a faster turnover of the globular head of myosin moving along the thin filament. T3-induced increases in the transcription of the MHC ~ gene have been demonstrated in whole animals [41] and in isolated neonatal rat myocytes [42]. This effect therefore is a direct consequence of thyroid hormone action and is independent of thyroid hormone-induced changes in the periphery. The marked thyroid hormone-induced changes in myosin isoenzyme predominance can be demonstrated in small animals (e.g., rats and rabbits) but do not occur to a similar extent in larger species, including the human heart. Changes in thyroid status also lead to marked alterations in the duration of diastolic relaxation [43] in addition to the changes in the speed of systolic contraction. Alterations in diastolic relaxation have been related to changes in the activity of the calcium ATPase pump of the SR [44-46]. The calcium ATPase of the SR is an ion pump that is responsible for removing calcium from the cytosol and sequestering it in the membranous structures of the SR during diastole. The resultant lowering of the systolic calcium concentration in diastole leads to the relaxation of cardiac muscle. Changes in calcium ATPase activity could occur through alterations in the number Of pump units or in the enzymatic activity and kinetic behavior of the pump enzyme. To gain insight into this mechanism, we quantitated the mRNA coding for the calcium ATPase of the SR. This quantitation was performed on Northern blots using specific cDNA probes. Quantitation of mRNA levels indicated that T3 administration to hypothyroid rats markedly and rapidly increases the mRNA coding for the slow isoform of the calcium ATPase in the heart [47]. Recent evidence indicates that the T3-induced increase in SR calcium ATPase mRNA levels results from increased transcription of the SR calcium ATPase gene [48]. It therefore appears that thyroid hormone increases the number of SR calcium ATPase pump units by increasing the transcription of the gene for this important contractile protein. In addition to MHC a, MHC fl, and
June 1990 The American Journal of Medicine Volume 88
Ta ACTION IN THE HEART / DILLMANN
the SR calcium ATPase, other T3-responsive mRNAs have been identified in the heart (Table II). Thyroid hormone-induced increases in the mRNA coding for malic enzyme [49], sodium-potassium ATPase [50], and the atrial natriuretic factor [51] have been documented. Furthermore, it has been shown that the number of calcium channels [52] increases, leading to increased calcium influx across the sarcolemma; however, it is currently unclear if this effect occurs through increases in the level of the specific mRNA. Similarly, increases in the number of ~ receptors have been documented [53]. The stimulatory effects that thyroid hormone has on the expression on the MHC a gene and on the gene coding for SR calcium ATPase lead to a positive inotropic effect in the heart. The questions therefore can be raised as to why thyroid hormone does not serve as a positive inotropic agent and why hyperthyroidism can promote cardiac failure [54]. One has to note that increases in the V1 form of myosin and increases in the pump numbers of SR calcium ATPase lead to a marked increase in ATP consumption by the heart. To determine how changes in thyroid status influence ATP consumption and the use of the chemical energy stored in ATP, the following studies were undertaken by Alpert and Mulieri [55]. Papillary muscle was obtained from rabbits made hyperthyroid, normal rabbits, and rabbits that had cardiac hypertrophy. In these papillary muscle preparations, the amount of chemical energy of ATP used for useful contractile movement and the amount released as heat were determined. The findings indicated that a larger fraction of the chemical energy of ATP was converted into heat in papillary muscle obtained from hyperthyroid hearts in comparison to normal papillary muscle. This inefficient use of energy may provide an explanation for the clinically appreciated finding that hyperthyroidism of long duration and great severity can lead to cardiac failure.
REFERENCES 1. Williams GH, Braunwald E: Endocrine and nutritional disorders and heart disease. In: Braunwald E, ed. Heart disease: a textbook of cardiovascular medicine, vol 2. Philadelphia: WB Saunders, 1988; 1800-1827. 2. Parry CH: Collections from the unpublished papers of the late Caleb Hilliel Parry, vol 2. London: 1825; 111. 3. Graves RJ: Clinical lectures. Lend Med Surg J (part 2) 1835; 7: 516. 4. Zak R, Martin AF, Reddy MK, Rabinowitz M: Control of protein balance in hypertrophied cardiac muscle. Circ Res 1976; 38(part 1): 145-150. 5. Rao GS, Eckel J, Rao ML, Breuer H: Uptake of thyroid hormone by isolated rat liver cells. Biochem Biophys Res Commun 1976; 73: 98-104. 6. Krenning EP, Doctor R, Bernard HF, Visser TJ, Hennemann G: Active transport of triiodothyronine (T3) into isolated rat liver cells (letter). FEBS Lett 1978; 91: 113. 7. Oppenheimer JH, Schwartz HL: Stereospecific transport of triiodothyronine from plasma to cytosol and from cytosol to nucleus in rat liver, kidney, brain, and heart. J Clin Invest 1985; 75: 147-154. 8. Pontecorvi A, Lakshmanan M, Robbins J: Different intraceliular and intranuclear transport of triiodothyronine enantiomers in rat skeletal myoblasts. Endocrinology 1988; 123: 2922-2929. 9. Schwartz HL, Trence D, Oppenheimer JH, Jiang NS, Jump DB: Distribution and metabolism of L- and D-triiodothyronine (T3) in the rat: preferential accumulation of L-T3 by hepatic and cardiac nuclei as a probable explanation of the differential biological potency of T3 enantiomers. Endocrinology 1983; 113: 1236-1243. 10. Segal J, Schwartz H, Gordon A: The effect of triiodothyronine on 2-deoxy-D-[1aH] glucose uptake in cultured chick embryo heart cells. Endocrinology 1977; 101: 143-149. 11. Mylotte KM, Cody V, Davis PJ, Davis FB, Bias SD, Schoenl M: Milrinone and thyroid hormone stimulate myocardial membrane Ca2+-ATPaseactivity and share structural homologies. Proc Natl Acad Sci USA 1985; 82: 7974-7978. 12. Sterling K, Campbell GA, Brenner MA: Purification of the mitochondrial triiodothyronine (T3) receptor from rat liver. Acta Endocrinol 1984; 105: 391-397. 13. Sanferd CF, Griffin EE, Wildenthal K: Synthesis and degradation of myocardial protein during the development and regression of thyroxine-induced cardiac hy-
pertrophy in rats. Circ Res 1978; 43: 688-694. 14. Oppenheimer JH, Koerner D, Schwartz HL, Surks MI: Specific nuclear triiodothyronine binding sites in rat liver and kidney. J Clin Endocrinol Metab 1972; 35: 330-333. 15. Samuels HH, Tsai JS, Casanova J, Stanley F: Thyroid hormone action: in vitro characterization of solubilized nuclear receptors from rat liver and cultured GH1 cells. J Clin Invest 1974; 54: 853-865. 16. Koerner D, Schwartz HL, Surks MI, Oppenheimer JH, Jorgensen EC: Binding of selected iodothyronine analogues to receptor sites of isolated rat hepatic nuclei: high correlation between structural requirements for nuclear binding and biological activity. J Biol Chem 1975; 250: 6417-6423. 17. Sap J, Munoz A, Damm K, et at The c-erb-A protein is a high-affinity receptor for thyroid hormone. Nature (Lend) 1986; 324: 635-640. 18. Weinberger C, Thompson CC, Ong ES, Lebo R, Groul DJ, Evans RM: The c-erbA gene encodes a thyroid hormone receptor. Nature (Lend) 1986; 324: 641-646. 19. Weinberger C, Hollenberg SM, Rosenfeld MG, Evans RM: Domain structure of human glucocorticoid receptor and its relationship to the v-erb-A oncogene product. Nature 1985; 318: 670-672. 20. Mitsuhashi T, Tennyson GE, Nikodem VM: Alternative splicing generates messages encoding rat c-erb-A proteins that do not bind thyroid hormone. Proc Natl Acad Sci USA 1988; 85: 5804-5808. 21. Thompson CC, Weinberger C, Lebo R, Evans RM: Identification of a novel thyroid hormone receptor expressed in the mammalian central nervous system. Science 1987; 237: I610-1614. 22. Benbrook D, Pfahl M: A novel thyroid hormone receptor encoded by a cDNA clone from a human testis library. Science 1987; 238: 788-791. 23. Lazar MA, Hodin RA, Darling DS, Chin WW: Identification of a rat c-erbA~related protein which binds deoxyribonucleic acid but does to bind thyroid hormone. Mol Endocrinol 1988; 2: 893-901. 24. Nakai A, Sakurai A, Bell GI, DeGree] LJ: Characterization of a third human thyroid hormone receptor coexpressed with other thyroid hormone receptors in several tissues. Mol Endocrinol 1988; 2: 1087-1092. 25. Hodin RA, Lazar MA, Wintman BI, ef ak Identification of a thyroid hormone receptor that is pituitary-specific. Science 1989; 244: 76-79. 26. Lazar MA, Hodin RA, Darling DS, Chin WW: A novel member of the thyroid/ steroid hormone receptor family is encoded by the opposite strand of the rat c-erb Aa transcriptional unit. Mol Cell Bio[ 1989; 9: 1128-1136. 27. Miyajima N, Horiuchi R, Shibuya Y, etal.'Two erbA homologs encoding proteins with different T3 binding capacities are transcribed from opposite DNA strands of the same genetic locus. Cell 1989; 57: 31-39. 28, Murray MB, Zilz ND, McCreary NL, MacDonald MJ, Towle HC: Isolation and characterization of rat cDNA clones for two distinct thyroid hormone receptors. J Biol Chem 1988; 263: 12770-12777. 29. Damm K, Thompson CC, Evans RM: Protein encoded by v-erbA functions as a thyroid-hormone receptor antagonist. Nature 1989; 339: 593-597. 30. Sap J, Munoz A, Schmitt J, Stunnenberg 14, Vennstrom B: Repression of transcription mediated at a thyroid hormone response element by the v-erb-A oncogene product. Nature 1989; 340: 242-244. 31, Evans RM: The steroid and thyroid hormone receptor superfamily. Science 1988; 240: 889-895. 32. Shapiro LE, Samuels HN, Yaffe BM: Thyroid and glucocerticoid hormones synergistically control growth hormone mRNA in cultured GH] cells. Proc Natl Acad Sci USA I978; 75: 45-49. 33. Dobner PR, Kawasaki EW, Yu L-Y, Bancroft FC: Thyroid or glucocorticoid hormone induces pre-growth-hormone mRNA and its probable nuclear precursor in rat pituitary cells. Proc Natl Acad Sci USA 1981; 78: 2230-2234. 34. Bedo G, Santisteban P, Aranda A: Retinoic acid regulates growth hormone gene expression. Nature 1989; 339: 231-234. 35. Umesono K, Giguere V, Glass CK, Rosenfeld MG, Evans RM: Retinoic acid and thyroid hormone induce gene expression through a common responsive element. Nature 1988; 336: 262-265. 36. Morita S, Fernandez-Mejia C, Melmed S: Retinoic acid selectively stimulates growth hormone secretion and messenger ribonucleic acid levels in rat pituitary cells. Endocrinology 1989; 124: 2052-2056. 37. Buccino RA, Spann JF, Peele PE, Braunwald E: Influence of the thyroid state on the intrinsic contractile properties and the energy stores of the myocardium. J Clin Invest 1967; 46: 1669-1682. 38. Emerson CP Jr, Bernstein SI: Molecular genetics of myosin. Annu Rev Biochem 1987; 56: 695-726. 39. Hoh JFY, McGrath PA, Hale PT: Electrophoretic analysis of multiple forms of cardiac myosin: effect of hypophysectomy and thyroxine replacement. J Mol Cell Cardiol 1978; 10: 1053-1076. 40. Schwartz K, Lecarpentier Y, Martin JL, et al: Myosin isoenzyme distribution correlates with speed of myocardial contraction. J Mol Cell Cardiol 1981; 13: 1017-1075. 41. Umeda PK, Levin JE, Sinha AM, et ak Molecular anatomy of cardiac myosin heavy chain genes. In: Emerson C, Fischman D. NadaI-Ginard B, Siddiqui MAQ, eds. Molecular biology of muscle development, UCLA Symposium on Molecular and Cellular Biology New Series, vol. 29. New York: Alan R Liss, 1986; 809-823. 42. Gustafson TA, Bahl JJ, Markham BE, Roeske WR, Morkin E: Hormonal regulation of myosin heavy chain and ~-actin gene expression in cultured fetal rat heart myocytes. J Biol Chem 1987; 262: 13316-13322. 43. Taylor RR, Covell JW, RossJ Jr: Influence of the thyroid state on left ventricular
June 1990
The American Journal of Medicine
Volume 88
629
T3 ACTION IN THE HEART / DILLMANN tension-velocity relations in the intact, sedated dog. J Clin Invest 1969; 48: 775784. 44. Suko J: Alterations of Ca2+-uptakeand Ca2+-activatedATPaseof cardiac sarcoplasmic reticulum in hyper- and hypothyroidism. Biochem BiophysActa 1971; 252: 324-327. 45. Nayler WG, Merrillees NCR, Chipperfield D, Kurtz J8: Influence of hyperthyroidism on the uptake and binding of calcium by cardiac microsomal fractions and on mitochondrial structure. Cardiovasc Res ]971; 5: 569-582. 46. Limas CJ: Calcium transport ATPase of cardiac sarcoplasmic reticulum in experimental hyperthyroidism. Am J Physiol 1978; 235: H745-H751. 47. Rohrer D, Dillmann WH: Thyroid hormone markedly increasesthe mRNAcoding for sarcoplasmic reticulum Ca2+-ATPasein the rat heart. J Biol Chem ]988; 263: 6941-6944. 48. Rohrer DK, Mestril R, Sayen R, Dillmann WH: Thyroid hormone induced increasesin rat heart sarcoplasmic reticulum Ca++ ATPasemRNAare transcriptionally mediated and require serum factors. Endocrinology 1989; 122(suppl): T-87. 49. Dozin B, Manguson MA, Nikodm VM: Thyroid hormone regulation of malic
630
June 1990 The American Journal of Medicine Volume88
enzyme synthesis. J Biol Chem 1986; 261: 10290-10292. 50. Gick GG, IsmaiI-Beigi F, Edelman IS: Thyroidal regulation of rat renal and hepatic Na, K-ATPasegene expression. J Biol Chem 1988; 263: 1661016618. 51. LadensonPW, Bloch KD, SeidmanJG: Modulationof atrial natriuretic factor by thyroid hormone: messenger ribonucleic acid and peptide levels in hypothyroid, euthyroid, and hyperthyroid rat atria and ventricles. Endocrinology 1988; 123: 652-657. 52. Kim D, Smith TW, Marsh JD: Effect of thyroid hormone on slow calcium channel function in cultured chick ventricular cells. J Clin Invest 1987; 80: 88-94. 53. Stiles GL, Lefkowitz RJ: Thyroid hormone modulation of agonist-betaadrenergic receptor interactions in the rat heart. Life Sci 1981; 28: 2529-2536. 54. CavalloA, Joseph CJ, Casta A: Cardiac complications in juvenile hyperthyroidism. Am J Dis Child ]984; ]38: 479-482. 55. Alpert NR, Mulieri LA: Thermomechanical economy of hypertrophied hearts. In: Alpert NR, ed. Perspectivesin cardiovascular research: myocardial hypertrophy and failure. New York: Raven, 1983; 619-630.
M.D., San Diego, Cafifornia
Thyroid hormone-induced changes in cardiac function have been recognized for over 150 years; however, the biochemical basis of triiodothyronine (T3) action in the heart hA~ been intensely investigated only during the last two decades. T3=induced changes in cardiac function can result from direct or indirect T3 effects. Direct T3 effects result from T3 action in the heart itself and are mediated by nuclear or extranuclear mechanisms. Extranuclear T3 effects, which occur independent of nuclear T3 receptor binding and increases in protein synthesis, influence primarily the transport of amino acids, sugars, and calcium across the cell membrane. Nuclear T3 effects are mediated by the binding of T3 to specific nuclear receptor proteins, which results in increased transcription of T3-responsive cardiac genes. The T3 receptor is a member of the ligandactivated trRnseription factor family and is encoded by cellular erythroblastosis A (c-erb A) genes. The c-erb A protein is the cellular homologue of the viral erythroblastosis A (v-erb A) protein, which causes red cell leukemi~ in chickens. Currently, three T3-binding isoforms of the c-erb protein and two non-T3-binding nuclear proteins that exert positive and negative effects on T3-responsive cardiac genes have been identified. T3 increases the heart trRn~cription of the myosin heavy cheln (MHC) a gene and decreases the transcription of the MHC ~ gene, leading to an increase of myosin VI and a decrease in myosin V3 isoenzymes. Myosin V], which is composed of two MHC ~, has a higher myosin ATPase activity than myosin V3, which cont~i-~ two MHC ~. The globular head of myosin V1, with its higher ATPase activity, leads to a more rapid movement of the glob, i~r head of myosin along the thin filament, resulting in an increased velocity of contraction. T3 also leads to an increase in the speed of diastolic relaxation, which is caused by the more efficient pumping of the calcium ATPase of the sarcoplasmic reticulum (SR). This T3 effect results from T3-induced increases in the level of the mRNA coding for the SR calcium ATPase protein, leading to an increased number of calcium ATPase pump units in the SR. Overall, thyroid hormone leads to an increase in ATP consumption in the heart. In addition, less chemical energy of ATP is used for contractile purposes and more of it goes toward heat production, which causes a decreased efficiency of the contractile process in the hyperthyroid heart. From the Department of Medicine, Division of Endocrinology and Metabolism, University of California, San Diego, San Diego, California. This work was supported by Grant HL25022 from the National Institutes of Health, Bethesda, Maryland, and was presented in part at the 64th Annual Meeting of the American Thyroid Association, San Francisco, California. Requests for reprints should be addressed to W.H. Dillmann, M.D., University of California at San Diego Medical Center, 225 Dickinson Street (H-811-C), San Diego, California 92103.
626
June 1990 The American Journalof Medicine
Volume 88
he heart is a major target organ for thyroid horT mone action, and marked changes occur in cardiac function in patients with hypothyroidism or hyperthyroidism [1]. This was first recognized in 1785 by Caleb Parry [2], a physician from the English resort town of Bath, who noted an association between swelling of the thyroid area and heart failure. He stated "there is one malady which I have in five cases seen coincidence which appears to be an enlargement of the heart. The malady to which I elude is enlargement of the thyroid gland." Fifty years later Robert Graves [3] described three cases of violent and long continued palpitation in women with thyrotoxicosis. Many important studies related to physiologic alterations induced by changes in the thyroid status were conducted subsequently; however, a detailed elucidation of the biochemical basis of thyroid hormone effects in the heart occurred only during the last 2 decades. It was also appreciated that thyroid status-induced changes in cardiac function can result from direct cardiac effects of thyroid hormone, that is, action of triiodothyronine (T3) itself in the cell of the heart versus indirect T3 effects, which are mediated by alterations occurring in the periphery resulting in hemodynamic changes that lead to secondary alterations in cardiac function. This discussion will be focused on direct effects of thyroid hormone on the heart. In considering thyroid hormone effects on cardiac tissue, it is important to emphasize that cardiac myocytes, which due to their large size contain the majority of cardiac protein and ribonucleic acids, constitute only one-third of the total cells that make up the myocardium [4]. The majority of cardiac cells are constituted by fibroblasts, smooth muscle cells, endothelial cells, and other cell types. Thyroid hormone-responsive proteins that have been investigated are primarily of myocytic origin, and thyroid hormone influences on nonmyocytic cardiac cells have not been investigated in detail. The first step in the sequence of events leading to thyroid hormone action in cardiac cells involves the transport of thyroid hormone across the plasma membrane or sarcolemma of cardiac myocytes, and subsequently transport across the nuclear membranes into the cell nucleus occurs. Recent findings by several groups of investigators indicate that an energy-requiring stereospecific transport step of thyroid hormone and its an~ogues occurs at the level of the plasma and nuclear membrane [5-8]. The stereospecific transport may account for discrepancies between nuclear thyroid hormone receptor binding and biologic activity. It has, for example, been noted that the binding of the LT3 isomer for the nuclear T3 receptor protein is only 15% to 20% higher than that of the D-T3 isomer [9]. In contrast, the biologic activity of D-T3 is only one third of L-T3. An explanation for this difference in receptor binding affinity and biologic activity may be provided by the preferential transport of L-T3 versus D-T3 across the plasma and nuclear membrane in the heart
T3 ACTION IN THE HEART / DILLMANN
[7]. Similar observations have been made in a rat myoblast cell line [8]. Once thyroid hormones enter cardiac cells, they mediate their effects through either a nuclear mechanism by binding to specific nuclear T3 receptors and stimulation of protein synthesis or by socalled extranuclear effects, which occur independent of binding to specific nuclear T3 receptor proteins and do not require ongoing protein synthesis. Extranuclear effects resulting in rapid stimulation of amino acid and sugar transport have been demonstrated [10]. In addition, thyroid hormone seems to exert an extranuclear effect on the sarcolemmal calcium ATPase, which results in increased calcium efflux from myocytes [11]. The potency of specific thyroid hormone analogues mediating this effect does not correspond to that observed for nuclear action. For example, T3 is almost as potent as thyroxine (T4) in stimulating sarcolemmal calcium ATPase [11]. In addition, direct effects of thyroid hormone on mitochondrial oxidative phosphorylation have been postulated [12]. Thyroid hormone effects on protein synthesis result in enhanced formation of total cardiac proteins [13] and, in addition, increases in the synthesis rate of the specific proteins that exceed the general increase in protein synthesis induced by thyroid hormone. For other specific proteins like myosin heavy chain (MHC) fl, the synthesis rate may decrease. T3-induced changes in the formation of specific proteins can result from alterations at the following steps. Thyroid hormone can (1) influence the formation of specific proteins by increasing the transcription of a given gene; (2) control how the primary transcript is spliced or otherwise processed; (3) select which mRNA species leave the nucleus and are exported to the cytoplasm; (4) selectively degrade certain mRNAs in the cytoplasm, influencing the half-life of the mRNA; (5) determine which mRNA species in the cytoplasm translate p r e f e r e n t i a l l y by ribosomes, l e a d i n g to translational control; and (6) selectively activate or degrade specific protein molecules after they have been made, leading to protein activity control. Thyroid hormone-induced changes in the formation of specific proteins appear to be controlled primarily at the steps regulating the transcription of specific genes and by influences on the stability of the mRNA in the cytoplasm. Thyroid hormone-induced changes in transcription are initiated by the binding of T3 to nuclear T3 receptors. Evidence for the existence of such receptors was first presented in 1972 and 1973 [14,15]. T3 receptor proteins are acidic nuclear proteins with a molecular weight of roughly 50,000 dalton (d) and bind thyroid hormone with high affinity. Because T3 is bound to this receptor protein with a higher affinity than T4, this thyroid hormone analogue has a higher biologic activity [16]. The following model for nuclear T3 effects was therefore developed. T3 enters the cell and the nucleus and is bound to the specific T3 receptor protein that is always present in association with chromatin. By binding of T3 to these receptor proteins, the transcription of specific genes is activated and an ificreasing amount of specific mRNA is formed. Attempts to purify T3 receptor proteins by conventional methods were unsuccessful because the protein was present at a low level (5,000 to 10,000 receptors/cell). In addition, the protein is labeled and difficult to puri-
TABLE I c.erb A Nuclear Proteins (T3 Receptors)*
Gene
Isoforms
c-erb Aa /
Protein
I
V/
I
I ,////I
N
~ o ~ =
T3 Effect
E
C
I' ~, GH
E
E c-erb Aft__......_
N ,~2
[
I
t Pituitary
E c-erbAreverse
--
v-erbA
--
N I
C I"1////I E
N I
C I /.///I E
,~GH
N = amino-terminal portion of the protein; C = carboxy-terminal portion of the protein; E = E domain of the receptor protein where T3 binds;" = increase in gene transcription; = decrease in gene transcription; GH = growth hormone. * Different c-erb A nuclear proteins are listed in this table. They are encoded by two different genes, c-erb A~ and c-erbA/~. The splicing variants c-erb Ao~2 and c-erb A/~2 arise from the o~ gene and c-erb A~2 does not bind T3 in the E domain of the receptor protein. The c-erbAo~2protein inhibits T3 effects on growth hormone gene transcription.
fy away from other proteins in the 40,000 to 60,000-d molecular weight range, which are of much higher predominance. Further characterization and identification of the nuclear T3 receptors as members of the cellular erythroblastosis (c-erb) A proto-oncogene family did not occur until 15 years later [17,18]. The first step in this identification process was the recognition that the amino acid sequence of the DNA binding region of the glucocorticoid receptor had a 47% homology to the viral erythroblastosis (v-erb) A protein [19]. This protein is a product from an oncogene virus that in conjunction with the v-erb B protein induces a red cell leukemia in chickens. The cellular homologue of this protein c-erb A had a molecular weight of 50,000 d and had the characteristics of a hormone-binding nuclear protein. It did not bind cortisol or other steroid hormones but bound T3 with similar affinities to those of the naturally occurring T3 receptor. The nuclear T3 receptor was therefore identical to the c-erb A protein. In addition, it could be shown that several isoforms of the T3 receptor-c-erb A proteins occur (Table I). At least two different genes, one coding for a c-erb Aa T3 receptor isoform and a second gene coding for the cerb A~ isoform, have been initially identified [17,18]. The transcriptional product of the c-erb Aa gene is further processed into the c-erb Aalfr3 receptor, which binds T3 and has a positive effect on the transcription of the growth hormone gene, and an alternatively spliced variant c-erb Aa2 is formed in cells, which does not bind T3 [20-24]. This isoform inhibits the transcription of the growth hormone gene but does not seem to have an influence on the transcription of the MHC a gene. The c-erb AB transcript is processed into a f~l and a f12form. The c-erb A~2 variant occurs in the pituitary and cannot be found in the heart [25]. The DNA region that codes for the c-erb Aa gene is
June 1990 The American Journal of Medicine Volume 88
627
T3 ACTION IN THE HEART/ DILLMANN TABLE II T3-Responsive mRNA in the Heart Type
MHCo
MHC fl SR calcium ATPase Malic enzyme
Gene
mRNA
;
t
Protein
Activity
t My°s'nAzPaseI Myosin ATPase Calcium sequestration 1.
Sodium efflux 1. ATPase ANF Calcium channel fl Receptor
;
1.
1'
?
?
?
--
--
1.
Sodium excretion 1. Calcium influx ~' Sympathetic stimulus 1.
ANF = atrial natriuretic factor; 1' = increase in level; J, = decrease in level; NAD PH = nicotinamide adenine dinucleotide phosphate hydrogen..
also transcribed in the reverse direction, leading to the c-erb Aa reverse protein, which does not bind T3 [26,27]. This DNA region can therefore lead to three isoforms of the c-erb Aa T3 receptor protein: the ~1 form, which binds T3, the ~2 form, which does not bind T3, and the reverse isoform, which also does not bind to T3. In cardiac tissue, the mRNA coding for c-erb Aal, c-erb A~2, c-erb Aflb and c-erb A reverse can be documented [26,28]. The v-erb A proto-oncogene also has an inhibitory effect on the action of the c-erb A~I T3 receptor [29,30]. The different isoforms of the T3 receptor and the nuclear receptor for steroid hormones are from a family of ligand-activated transcriptional factors that show a significant degree of homology [31]. All hormone ligands contain a ring structure made up of six carbon atoms and are lipophilic and hydrophobic. The ligand binds the nuclear receptor protein at the carboxyterminal end. The receptor protein contains a DNA binding region (C region) in which these different receptors show a high degree of homology. The T3 receptor proteins bind with the C-region to areas in the DNA containing nucleotide consensus sequences that are localized upstream from the origin of transcription of T3-responsive genes. It is of interest to note that, for example, the growth hormone gene is thyroid hormone-responsive and also responds to glucocorticoids and to retinoic acid [32-36]. One could Speculate that during evolution, a primordial six-carbon ligand and a corresponding binding protein arose and were then modified into the specific receptors for thyroid hor: mone, glucocorticoids, retinoic acid, and other ligands. In addition, specific genes remained responsive to those ligands and their receptor proteins. The following model of T3 receptor action arises from recent findings. It appears that the T3 receptor is always associated with DNA in areas upstream from the transcription initiation site of thyroid hormone-responsive genes. The unoccupied T3 receptor acts as a transcriptional inhibitor, repressing the transcription of the thyroid hormone-responsive gene. Association of T3 with the carboxy-terminal E domain of the T3 receptor leads to a change in the confirmation of the receptor protein, which relieves the repressive function and allows the gene to be transcribed. Illustrative of how T3-induced changes in the tran-
628
scription of specific genes can result in contractile alterations are the T3 effect on the MHC ~ gene and the sarcoplasmic reticulum (SR) calcium ATPase gene. We will first consider alterations in MHC gene expression. It has been known for some time that increases in thyroid hormone levels lead to increases in the velocity of systolic contraction of the heart [37]. These alterations can be explained by alterations in myosin isoenzyme predominance. The myosin holoenzyme has a molecular weight of approximately 500,000 d and consists of two MHCs and four light chains [38]. Myosin V1, which predominates in the normal rat heart, consists of two MHC ~, whereas myosin V3 contains two MHC/3, and myosin V2 is a heterodimer of MHC ~ and MHC ~. Myosin V1 has markedly higher myosin ATPase activity than myosin V3. In the hypothyroid rat heart, myosin V3 predominates and myosin with low ATPase activity participates in the contractile process, leading to a decreased velocity of contraction in hypothyroid papillary muscle [39,40]. In contrast, thyroid hormone administration to hyperthyroid rats leads to increased expression of the MHC ~ gene, resulting in a markedly higher fraction of myosin isoenzymes being composed of two MHC a. The higher myosin ATPase activity of myosin V1 leads to a faster turnover of the globular head of myosin moving along the thin filament. T3-induced increases in the transcription of the MHC ~ gene have been demonstrated in whole animals [41] and in isolated neonatal rat myocytes [42]. This effect therefore is a direct consequence of thyroid hormone action and is independent of thyroid hormone-induced changes in the periphery. The marked thyroid hormone-induced changes in myosin isoenzyme predominance can be demonstrated in small animals (e.g., rats and rabbits) but do not occur to a similar extent in larger species, including the human heart. Changes in thyroid status also lead to marked alterations in the duration of diastolic relaxation [43] in addition to the changes in the speed of systolic contraction. Alterations in diastolic relaxation have been related to changes in the activity of the calcium ATPase pump of the SR [44-46]. The calcium ATPase of the SR is an ion pump that is responsible for removing calcium from the cytosol and sequestering it in the membranous structures of the SR during diastole. The resultant lowering of the systolic calcium concentration in diastole leads to the relaxation of cardiac muscle. Changes in calcium ATPase activity could occur through alterations in the number Of pump units or in the enzymatic activity and kinetic behavior of the pump enzyme. To gain insight into this mechanism, we quantitated the mRNA coding for the calcium ATPase of the SR. This quantitation was performed on Northern blots using specific cDNA probes. Quantitation of mRNA levels indicated that T3 administration to hypothyroid rats markedly and rapidly increases the mRNA coding for the slow isoform of the calcium ATPase in the heart [47]. Recent evidence indicates that the T3-induced increase in SR calcium ATPase mRNA levels results from increased transcription of the SR calcium ATPase gene [48]. It therefore appears that thyroid hormone increases the number of SR calcium ATPase pump units by increasing the transcription of the gene for this important contractile protein. In addition to MHC a, MHC fl, and
June 1990 The American Journal of Medicine Volume 88
Ta ACTION IN THE HEART / DILLMANN
the SR calcium ATPase, other T3-responsive mRNAs have been identified in the heart (Table II). Thyroid hormone-induced increases in the mRNA coding for malic enzyme [49], sodium-potassium ATPase [50], and the atrial natriuretic factor [51] have been documented. Furthermore, it has been shown that the number of calcium channels [52] increases, leading to increased calcium influx across the sarcolemma; however, it is currently unclear if this effect occurs through increases in the level of the specific mRNA. Similarly, increases in the number of ~ receptors have been documented [53]. The stimulatory effects that thyroid hormone has on the expression on the MHC a gene and on the gene coding for SR calcium ATPase lead to a positive inotropic effect in the heart. The questions therefore can be raised as to why thyroid hormone does not serve as a positive inotropic agent and why hyperthyroidism can promote cardiac failure [54]. One has to note that increases in the V1 form of myosin and increases in the pump numbers of SR calcium ATPase lead to a marked increase in ATP consumption by the heart. To determine how changes in thyroid status influence ATP consumption and the use of the chemical energy stored in ATP, the following studies were undertaken by Alpert and Mulieri [55]. Papillary muscle was obtained from rabbits made hyperthyroid, normal rabbits, and rabbits that had cardiac hypertrophy. In these papillary muscle preparations, the amount of chemical energy of ATP used for useful contractile movement and the amount released as heat were determined. The findings indicated that a larger fraction of the chemical energy of ATP was converted into heat in papillary muscle obtained from hyperthyroid hearts in comparison to normal papillary muscle. This inefficient use of energy may provide an explanation for the clinically appreciated finding that hyperthyroidism of long duration and great severity can lead to cardiac failure.
REFERENCES 1. Williams GH, Braunwald E: Endocrine and nutritional disorders and heart disease. In: Braunwald E, ed. Heart disease: a textbook of cardiovascular medicine, vol 2. Philadelphia: WB Saunders, 1988; 1800-1827. 2. Parry CH: Collections from the unpublished papers of the late Caleb Hilliel Parry, vol 2. London: 1825; 111. 3. Graves RJ: Clinical lectures. Lend Med Surg J (part 2) 1835; 7: 516. 4. Zak R, Martin AF, Reddy MK, Rabinowitz M: Control of protein balance in hypertrophied cardiac muscle. Circ Res 1976; 38(part 1): 145-150. 5. Rao GS, Eckel J, Rao ML, Breuer H: Uptake of thyroid hormone by isolated rat liver cells. Biochem Biophys Res Commun 1976; 73: 98-104. 6. Krenning EP, Doctor R, Bernard HF, Visser TJ, Hennemann G: Active transport of triiodothyronine (T3) into isolated rat liver cells (letter). FEBS Lett 1978; 91: 113. 7. Oppenheimer JH, Schwartz HL: Stereospecific transport of triiodothyronine from plasma to cytosol and from cytosol to nucleus in rat liver, kidney, brain, and heart. J Clin Invest 1985; 75: 147-154. 8. Pontecorvi A, Lakshmanan M, Robbins J: Different intraceliular and intranuclear transport of triiodothyronine enantiomers in rat skeletal myoblasts. Endocrinology 1988; 123: 2922-2929. 9. Schwartz HL, Trence D, Oppenheimer JH, Jiang NS, Jump DB: Distribution and metabolism of L- and D-triiodothyronine (T3) in the rat: preferential accumulation of L-T3 by hepatic and cardiac nuclei as a probable explanation of the differential biological potency of T3 enantiomers. Endocrinology 1983; 113: 1236-1243. 10. Segal J, Schwartz H, Gordon A: The effect of triiodothyronine on 2-deoxy-D-[1aH] glucose uptake in cultured chick embryo heart cells. Endocrinology 1977; 101: 143-149. 11. Mylotte KM, Cody V, Davis PJ, Davis FB, Bias SD, Schoenl M: Milrinone and thyroid hormone stimulate myocardial membrane Ca2+-ATPaseactivity and share structural homologies. Proc Natl Acad Sci USA 1985; 82: 7974-7978. 12. Sterling K, Campbell GA, Brenner MA: Purification of the mitochondrial triiodothyronine (T3) receptor from rat liver. Acta Endocrinol 1984; 105: 391-397. 13. Sanferd CF, Griffin EE, Wildenthal K: Synthesis and degradation of myocardial protein during the development and regression of thyroxine-induced cardiac hy-
pertrophy in rats. Circ Res 1978; 43: 688-694. 14. Oppenheimer JH, Koerner D, Schwartz HL, Surks MI: Specific nuclear triiodothyronine binding sites in rat liver and kidney. J Clin Endocrinol Metab 1972; 35: 330-333. 15. Samuels HH, Tsai JS, Casanova J, Stanley F: Thyroid hormone action: in vitro characterization of solubilized nuclear receptors from rat liver and cultured GH1 cells. J Clin Invest 1974; 54: 853-865. 16. Koerner D, Schwartz HL, Surks MI, Oppenheimer JH, Jorgensen EC: Binding of selected iodothyronine analogues to receptor sites of isolated rat hepatic nuclei: high correlation between structural requirements for nuclear binding and biological activity. J Biol Chem 1975; 250: 6417-6423. 17. Sap J, Munoz A, Damm K, et at The c-erb-A protein is a high-affinity receptor for thyroid hormone. Nature (Lend) 1986; 324: 635-640. 18. Weinberger C, Thompson CC, Ong ES, Lebo R, Groul DJ, Evans RM: The c-erbA gene encodes a thyroid hormone receptor. Nature (Lend) 1986; 324: 641-646. 19. Weinberger C, Hollenberg SM, Rosenfeld MG, Evans RM: Domain structure of human glucocorticoid receptor and its relationship to the v-erb-A oncogene product. Nature 1985; 318: 670-672. 20. Mitsuhashi T, Tennyson GE, Nikodem VM: Alternative splicing generates messages encoding rat c-erb-A proteins that do not bind thyroid hormone. Proc Natl Acad Sci USA 1988; 85: 5804-5808. 21. Thompson CC, Weinberger C, Lebo R, Evans RM: Identification of a novel thyroid hormone receptor expressed in the mammalian central nervous system. Science 1987; 237: I610-1614. 22. Benbrook D, Pfahl M: A novel thyroid hormone receptor encoded by a cDNA clone from a human testis library. Science 1987; 238: 788-791. 23. Lazar MA, Hodin RA, Darling DS, Chin WW: Identification of a rat c-erbA~related protein which binds deoxyribonucleic acid but does to bind thyroid hormone. Mol Endocrinol 1988; 2: 893-901. 24. Nakai A, Sakurai A, Bell GI, DeGree] LJ: Characterization of a third human thyroid hormone receptor coexpressed with other thyroid hormone receptors in several tissues. Mol Endocrinol 1988; 2: 1087-1092. 25. Hodin RA, Lazar MA, Wintman BI, ef ak Identification of a thyroid hormone receptor that is pituitary-specific. Science 1989; 244: 76-79. 26. Lazar MA, Hodin RA, Darling DS, Chin WW: A novel member of the thyroid/ steroid hormone receptor family is encoded by the opposite strand of the rat c-erb Aa transcriptional unit. Mol Cell Bio[ 1989; 9: 1128-1136. 27. Miyajima N, Horiuchi R, Shibuya Y, etal.'Two erbA homologs encoding proteins with different T3 binding capacities are transcribed from opposite DNA strands of the same genetic locus. Cell 1989; 57: 31-39. 28, Murray MB, Zilz ND, McCreary NL, MacDonald MJ, Towle HC: Isolation and characterization of rat cDNA clones for two distinct thyroid hormone receptors. J Biol Chem 1988; 263: 12770-12777. 29. Damm K, Thompson CC, Evans RM: Protein encoded by v-erbA functions as a thyroid-hormone receptor antagonist. Nature 1989; 339: 593-597. 30. Sap J, Munoz A, Schmitt J, Stunnenberg 14, Vennstrom B: Repression of transcription mediated at a thyroid hormone response element by the v-erb-A oncogene product. Nature 1989; 340: 242-244. 31, Evans RM: The steroid and thyroid hormone receptor superfamily. Science 1988; 240: 889-895. 32. Shapiro LE, Samuels HN, Yaffe BM: Thyroid and glucocerticoid hormones synergistically control growth hormone mRNA in cultured GH] cells. Proc Natl Acad Sci USA I978; 75: 45-49. 33. Dobner PR, Kawasaki EW, Yu L-Y, Bancroft FC: Thyroid or glucocorticoid hormone induces pre-growth-hormone mRNA and its probable nuclear precursor in rat pituitary cells. Proc Natl Acad Sci USA 1981; 78: 2230-2234. 34. Bedo G, Santisteban P, Aranda A: Retinoic acid regulates growth hormone gene expression. Nature 1989; 339: 231-234. 35. Umesono K, Giguere V, Glass CK, Rosenfeld MG, Evans RM: Retinoic acid and thyroid hormone induce gene expression through a common responsive element. Nature 1988; 336: 262-265. 36. Morita S, Fernandez-Mejia C, Melmed S: Retinoic acid selectively stimulates growth hormone secretion and messenger ribonucleic acid levels in rat pituitary cells. Endocrinology 1989; 124: 2052-2056. 37. Buccino RA, Spann JF, Peele PE, Braunwald E: Influence of the thyroid state on the intrinsic contractile properties and the energy stores of the myocardium. J Clin Invest 1967; 46: 1669-1682. 38. Emerson CP Jr, Bernstein SI: Molecular genetics of myosin. Annu Rev Biochem 1987; 56: 695-726. 39. Hoh JFY, McGrath PA, Hale PT: Electrophoretic analysis of multiple forms of cardiac myosin: effect of hypophysectomy and thyroxine replacement. J Mol Cell Cardiol 1978; 10: 1053-1076. 40. Schwartz K, Lecarpentier Y, Martin JL, et al: Myosin isoenzyme distribution correlates with speed of myocardial contraction. J Mol Cell Cardiol 1981; 13: 1017-1075. 41. Umeda PK, Levin JE, Sinha AM, et ak Molecular anatomy of cardiac myosin heavy chain genes. In: Emerson C, Fischman D. NadaI-Ginard B, Siddiqui MAQ, eds. Molecular biology of muscle development, UCLA Symposium on Molecular and Cellular Biology New Series, vol. 29. New York: Alan R Liss, 1986; 809-823. 42. Gustafson TA, Bahl JJ, Markham BE, Roeske WR, Morkin E: Hormonal regulation of myosin heavy chain and ~-actin gene expression in cultured fetal rat heart myocytes. J Biol Chem 1987; 262: 13316-13322. 43. Taylor RR, Covell JW, RossJ Jr: Influence of the thyroid state on left ventricular
June 1990
The American Journal of Medicine
Volume 88
629
T3 ACTION IN THE HEART / DILLMANN tension-velocity relations in the intact, sedated dog. J Clin Invest 1969; 48: 775784. 44. Suko J: Alterations of Ca2+-uptakeand Ca2+-activatedATPaseof cardiac sarcoplasmic reticulum in hyper- and hypothyroidism. Biochem BiophysActa 1971; 252: 324-327. 45. Nayler WG, Merrillees NCR, Chipperfield D, Kurtz J8: Influence of hyperthyroidism on the uptake and binding of calcium by cardiac microsomal fractions and on mitochondrial structure. Cardiovasc Res ]971; 5: 569-582. 46. Limas CJ: Calcium transport ATPase of cardiac sarcoplasmic reticulum in experimental hyperthyroidism. Am J Physiol 1978; 235: H745-H751. 47. Rohrer D, Dillmann WH: Thyroid hormone markedly increasesthe mRNAcoding for sarcoplasmic reticulum Ca2+-ATPasein the rat heart. J Biol Chem ]988; 263: 6941-6944. 48. Rohrer DK, Mestril R, Sayen R, Dillmann WH: Thyroid hormone induced increasesin rat heart sarcoplasmic reticulum Ca++ ATPasemRNAare transcriptionally mediated and require serum factors. Endocrinology 1989; 122(suppl): T-87. 49. Dozin B, Manguson MA, Nikodm VM: Thyroid hormone regulation of malic
630
June 1990 The American Journal of Medicine Volume88
enzyme synthesis. J Biol Chem 1986; 261: 10290-10292. 50. Gick GG, IsmaiI-Beigi F, Edelman IS: Thyroidal regulation of rat renal and hepatic Na, K-ATPasegene expression. J Biol Chem 1988; 263: 1661016618. 51. LadensonPW, Bloch KD, SeidmanJG: Modulationof atrial natriuretic factor by thyroid hormone: messenger ribonucleic acid and peptide levels in hypothyroid, euthyroid, and hyperthyroid rat atria and ventricles. Endocrinology 1988; 123: 652-657. 52. Kim D, Smith TW, Marsh JD: Effect of thyroid hormone on slow calcium channel function in cultured chick ventricular cells. J Clin Invest 1987; 80: 88-94. 53. Stiles GL, Lefkowitz RJ: Thyroid hormone modulation of agonist-betaadrenergic receptor interactions in the rat heart. Life Sci 1981; 28: 2529-2536. 54. CavalloA, Joseph CJ, Casta A: Cardiac complications in juvenile hyperthyroidism. Am J Dis Child ]984; ]38: 479-482. 55. Alpert NR, Mulieri LA: Thermomechanical economy of hypertrophied hearts. In: Alpert NR, ed. Perspectivesin cardiovascular research: myocardial hypertrophy and failure. New York: Raven, 1983; 619-630.
